197

11.1 Introduction

Lime mortar is a mixture formed by kneading together aggregates, lime and water. It is one of the most frequently used materials in the history of building and was widespread even in ancient times with remains being found in houses in Delos and Thera (Gaspar Tébar 1996 ), and in buildings in Festos and Malia (Furlan and Bissegger 1975 ; Malinowski 1981 ).

Its use however has varied at different times in history, and it almost went out of use when Portland cement appeared in 1824, because cement offered certain advan-tages such as fast setting and high mechanical resistance (De Buergo Ballester and González Limón 1994 ; Radonjic et al. 2001 ).

In recent years lime mortar has staged something of a comeback, and is used especially in restoration work on historical buildings (Bromblet 1999 ). There are two main reasons for this. Firstly, its high chemical and physical compatibility with the materials normally used in these buildings (Iglesias Martínez 1996 ; Pérez Mon-serrat and Baltuille Martín 2001 ), and secondly its mechanical properties, which make it capable of resisting some degree of movement in the masonry. By con-trast cement mortars, which in theory are stronger, are in fact less able to resist earthquakes (Hendry 2001 ). They are often incompatible with the materials used in historical buildings, and in some cases instead of repairing the damage to the building they have actually worsened it, causing decay that often proves irreversible (Arandigoyen and Alvarez 2007 ; Storemyr 2004 ). Another disadvantage of cement mortars is that salts from the alkalis normally contained in Portland cement have been shown to cause decay in the building stone (Hekal et al. 2002 ). The decline in the use of lime-based mortars was largely due to their slow carbonation which

M. Bostenaru Dan et al. (eds.), Materials, Technologies and Practice in Historic Heritage Structures, DOI 10.1007/978-90-481-2684-2_11 , © Springer Science+Business Media B.V. 2010

Chapter 11 The Use of Lime Mortars in Restoration Work on Architectural Heritage Ana Luque, Giuseppe Cultrone and Eduardo Sebastián

A. Luque ( ) Department of Mineralogy and Petrology University of Granada , Avda.Fuentenueva , 18002 Granada , Spain Tel.: +34-958-240077 Fax: +34-958-243368 e-mail: analuque@ugr.es

198 A. Luque et al.

meant that they took longer to harden than cement mortars. Carbonation is a natural process that makes mortars harder and therefore more durable (Lanas and Álvarez Galindo 2003 ). This process depends on many factors including CO 2 concentration, temperature and relative humidity (Dehilly et al. 2002 ; Martínez Ramírez et al. 2003 ) and normally involves an increase in weight caused by the transformation of portlandite into calcite (Moorehead 1986 , Dehilly et al. 2002 ). It has been observed that the total carbonation of mortar could take many years or even centuries (De La Torre 1995 ).

The study of lime mortars is now of great interest to restorers and conservation scientists involved in the safeguarding of our Architectural Heritage, however a bet-ter understanding of these materials is still required in order to promote their use.

Our objective in this work is to provide the restorer and researcher with a practi-cal guide to important aspects of lime mortars that can be applied in their work on our Architectural Heritage.

11.2 Preparation of Lime Mortars

Although the process of making a lime mortar is apparently quite straightforward, research has shown that careful control in the selection of materials and in the manufacturing process does help to produce a high quality mortar. The choice of a particular aggregate or binder is therefore very important, as is the way the mixing process is carried out.

Nowadays there are a lot of regulations and standards that control the quality of the materials used in the kneading of lime mortars (UNE-EN, UNI-EN, ASTM, etc.). As far as the kneading process is concerned, however, although there are regu-lations, these tend to focus on the preparation of another kind of mortar, the cement mortar. Lime mortars and cement mortars have different properties which mean that these regulations cannot always be applied to the preparation of lime mortars (Cazalla 2002 ).

11.2.1 Selection of Materials

In ancient times, the choice of aggregate or binder was not subject to the same tight controls as today. This choice is however fundamental, as for example if a mortar is prepared with a lime that is not well fired or is partially hydrated, the result will be a poor quality mortar.

In the past the aggregate used by builders was generally taken from areas near the building site or very often from rivers. Today, however, there are specific rules as to its granulometry, the size of the grain, the porosity, surface, shape and the chemical and mineralogical composition. Of the aggregates currently available on the market, those of a siliceous composition are normally the most suitable. This

19911 The Use of Lime Mortars in Restoration Work on Architectural Heritage

type of aggregate was the one most commonly used in ancient times and shows great hardness and chemical resistance (De Buergo Ballester and Gonzalez Limón 1994 ), in addition to its good mechanical properties, durability and adherence.

As for the lime, this is a generic term used to describe all the physical forms in which calcium oxide (CaO), magnesium oxide (MgO), calcium hydroxide Ca(OH) 2 and magnesium hydroxide Mg (OH) 2 may appear. The UNE-EN 459-1/AC standard classifies limes on the basis of the environment in which they set and/or harden (they are also classified on the basis of their state in oxide or hydroxide form and their water content, Table 11.1 ) and on the basis of their chemical composition.

In general, and bearing in mind the use to which the mortars will be put, slaked lime of the air-slaked or fat lime varieties with a CaO content of over 90%, in either paste or powder form has been shown to be the best lime to use in the preparation of lime mortars (Elert et al. 2002 and references therein). There are also dolomitic limes in which the MgO content is over 10%. However this binder is much less frequently used (De La Torre et al. 1996 ). In our experience studying Muslim build-ings (9th to 16th centuries) for example we have never come across it. The lime they used was known as “old” or “aged” lime putty. This type of lime was used after submerging it in water (on rafts) for long periods of time.

Nowadays the product most widely manufactured is slaked lime in dry pow-der form (hydrated in a stechiometric way), while lime putty which is often better quality is very hard to find. This creates a problem when it comes to selecting the best lime to use to make the lime mortar. It is therefore very important to find out how these limes are worked and how similar they are to the “old” limes and what advantages and disadvantages they offer compared to the “old” limes once in place on the building.

11.2.2 Kneading Process

According to Cazalla ( 2002 ); Martín Pérez ( 1990 ); and Hoffman and Vetter ( 1990 ), the best binder-aggregate mix is 1:3, and this is also the most widely used in mortars for restoration work (Malinowski 1981 ; Sbordini Mora 1981 ). In our experience, however, if lime putty is used, this ratio can be higher (1:4 or 1:5) without the mor-tar losing any of its technical quality.

Opinions are divided as to the best order in which to add the materials. On the basis of our experience and depending on the form (“putty” or “dry powder”) in which the lime appears, we recommend adding the materials in the order set out

Table 11.1 Classification of limes according to UNE-EN 459-1/AC ( 2002 ) norms Firing temperature of raw material (ºC) Clay content (%) Air limes 800–900 5

Set in contact with air Hydraulic limes 1050–1150 5–22

Set in contact with air or with water

200 A. Luque et al.

in Table 11.2 . When working with limes in putty form, add the lime first, and then the aggregate and while these are being mixed together, gradually add the water in a very careful, controlled way. When working with limes in powder form, the lime must first be mixed with water and then kneaded, until a homogeneous paste with no lumps appears. Add the sand to this paste and continue kneading, so as to ensure that the final mortar paste is as evenly distributed and workable as possible (Ashurst 1990 ).

11.3 The Carbonation of Lime Mortars

11.3.1 Weight Increase Through Carbonation

The carbonation process that takes place in a mortar is caused by a reaction in which portlandite Ca(OH) 2 is transformed into calcite CaCO 3 . This process is influenced by the diffusion of CO 2 through the pores of the mortar, by the dissolution of CO 2 and Ca(OH) 2 in the water that condenses in the pores and by the chemical equilib-rium that is produced in this solution and the resulting precipitation of CaCO 3 (Van Balen and Van Gemert 1994 ).

Using X-Ray diffraction we can obtain valuable information about the degree of carbonation of a mortar, as we can measure the amount of calcite produced and then compare it to the amount of portlandite that remains after a certain period of time (R parameter defined by Cazalla et al. 2000 ). We do this by comparing the lines of maximum diffraction intensity for the calcite and the portlandite (2 = 29.400 and 2 = 34.033, respectively) (Fig. 11.1 ). During the first few days of curing, we could clearly observe how the peak showing the maximum intensity for portlandite was higher than the peak for calcite, while after a certain period of curing the peak for calcite was higher than that for portlandite. This shows that the carbonation process continues over time and can be considered to have come to an end when the dif-fractograms no longer show diffraction lines for portlandite.

It is important to understand the carbonation process, as the material is modified not only from a mineralogical point of view but also in a textural and mechanical sense. To help us to understand this process better we can simulate it in the labora-tory by subjecting the mortars to forced carbonation processes and then compare them with similar samples that have carbonated naturally.

We observed that lime mortars subjected to forced carbonation in a weather chamber with a temperature of 25ºC and a relative humidity of 50% registered a

Table 11.2 Changes in the sequence of materials we have to add during the kneading Lime putty Dry powder lime Sequence lime + sand + water lime + water + sand Dosage 1+3 + 0.5 or 0.9 1+1 + 3

20111 The Use of Lime Mortars in Restoration Work on Architectural Heritage

weight increase of 6–7% after 30 days. We also studied the behaviour of mortars with certain additives to see if these would in any way enhance the carbonation process. Mortars to which additives had been added (we studied the behaviour of lime mortars without additives and others with pozzolana and/or an air-entraining agent) obtained lower values. It is important to note that, after just eight days all the mortars (with or without additives) had already reached their maximum weight increase, and after that showed only small oscillations (Fig. 11.2a ).

In contrast, the weight increase of the mortars subjected to a natural process of carbonation was less than 0.5% after 20 days and after four months the figure was still less than 2%. If the weight of the mortar were to continue increasing at the rate recorded during the first days of natural carbonation, it would take one year to obtain the same results as those obtained with the samples carbonated in the climatic chamber (Fig. 11.2b ). Nevertheless, the rate of weight increase slowed down over time and entered an asymptotic curve when the weight increase was less than 3%, i.e., less than half the weight increase of the mortars subject to forced carbonation. After 6 months, the weight of the mortars was still rising, albeit very slowly.

Notice that all samples (forced and naturally carbonated) start to carbonate dur-ing the drying phase. The weight difference shown by these two groups of mortars after just 2 days of carbonation suggests that very little calcite would have been generated before the beginning of this test. Taking into account that temperature and relative humidity were the same for both forced and naturally carbonated sam-ples, the CO 2 concentration during the carbonation process appears to be a decisive parameter in lime mortar carbonation kinetics. The higher the CO 2 concentration,

Fig. 11.1 XRD diagrams of a lime mortar after one month and after 18 months of carbonation. Legend: Cc = calcite; P = portlandite

202 A. Luque et al.

the deeper the excess CO 2 can penetrate into the mortar block, thus producing a thicker carbonated area (leading to faster carbonation). At atmospheric CO 2 concen-tration, any CO 2 molecule entering the mortar pore system could rapidly react with Ca(OH) 2 . Thus all CO 2 molecules will be “trapped” near the surface of the mortar, before the reaction front progresses to the sample core. As long as there is unreacted Ca(OH) 2 on the surface layer, the carbonation front will not progress towards the core of the mortar.

The reaction rate is independent of CO 2 concentration. The rate depends on the reactivity of the lime (i.e., surface area) and the water content (Van Balen and Van Gemert 1994 ). However, even if we consider a constant reaction rate, the higher the CO 2 concentration, the faster, the more thorough the carbonation process is.

Bearing in mind the molecular weight of calcite (100.09 g) and portlandite (74.09 g) and the densities of portlandite (2.23 g cm 3 ) and α -quartz (2.53 g cm 3 ), and considering that in these mortars three parts are occupied by the aggregate ( α -quartz) and one by the lime, a weight increase of 7.91% could be expected if the portlandite was completely transformed into calcite. The 6% result obtained in forced carbonated mortars corresponds, therefore, to 75.85% carbonation of the initial mass of the lime. On the other hand, the 1.5% weight increase in naturally carbonated mortars indicates only 20.23% carbonation.

Fig. 11.2 Weight gain (in %) during forced a and natural carbonation b . 1 represents a mortar without additives, while 2 is a mortar with additive

3

8

6

4

2

0

6 8 10 13

daysa

b days

2 8 15 20 79 124 250 468

1

2

1

2

15 17 20 22 30M

/M (

%)

M/M

(%

)

8

6

4

2

0

20311 The Use of Lime Mortars in Restoration Work on Architectural Heritage

11.3.2 Calcium Carbonate Formation Through Carbonation

Figure 11.3 shows the graph of XRD analysis with regard to calcite concentra-tions in the mortars versus the time of carbonation for samples subjected to forced carbonation. There is a clear link with the results obtained by weight increase (cf. Fig. 11.2a ), and mortars without additives reach the highest degree of carbonation. After 48 h more than 50% of the portlandite had turned into calcite. After 6 days the percentage of calcite was nearly 65% and after 8 days it was over 90%. This shows that after 8 days in a CO 2 -saturated atmosphere almost all the mortar had carbon-ated, as weight increase data suggested. It is also clear that a 100% CaCO 3 value was not reached. The complete transformation of Ca(OH) 2 into CaCO 3 is difficult to achieve. This may be due to the heat that is produced during the transformation of portlandite into calcite, which in turn causes the capillary water inside the mor-tars to evaporate, especially when high CO 2 concentrations are present (Moorehead 1986 ). Van Balen ( 2005 ), however, demonstrated that the carbonation reaction of lime does not depend on the CO 2 concentration and that in fact the controlling fac-tors are the presence of water and the specific surface of lime. This hypothesis may explain why the reaction in forced carbonated samples ended after only 8 days, while in naturally carbonated mortars the process can last for years.

A second factor that can impede or at least reduce carbonation is the environ-mental temperature. Dehilly et al. ( 2002 ) observed the complete carbonation of portlandite, probably because of the lower temperatures they used in their investi-gation. It is a well known fact that the solubility of CO 2 decreases as temperature increases (Moorehead 1986 ). Another cause may be the reduction in porosity dur-ing carbonation as a result of calcite crystallization, which reduces the space for gas molecules to migrate towards the calcium hydroxide crystals located inside the mortar (Stamatakis et al. 2001 ).

Therefore, the carbonation of mortar subjected to forced carbonation can start during the earlier drying phase, but contact with a CO 2 -rich atmosphere (in the pres-ence of water) is what accelerates the process. In fact, a comparison with the limited weight increase of samples subject to natural carbonation confirms this assertion.

Fig. 11.3 Percentage of newly-formed CaCO 3 in mortars during forced carbonation

2 6 8 10 13 15 17 20 22 26 30

days

CaC

O3

(%)

100

80

60

40

20

0

204 A. Luque et al.

11.4 Fabric and Physical Properties of Lime Mortars

11.4.1 Crystals Morphology and Fabric

Scanning Electron Microscopy allows us to study the microfabric of mortars, the shape and size of the pores and/or fissures and the degree of adherence between binder and aggregate.

Lime carbonation determines changes in the morphology of the crystals. At the beginning of the carbonation process a large number of hexagonal, plate-like crys-tals of portlandite are visible (Fig. 11.4a ). In some cases they are isolated, and in others they are heaped on top of each other. The size of these crystals can range from 200 nm to 1 μ m. During the carbonation process, portlandite platelets dis-appear and are replaced by calcite crystals of irregular morphology (Fig. 11.4b ). Finally, when the carbonation is almost complete, the mortar structure is covered by 1 μ m wide scalenohedral calcite crystals with a few remaining portlandite platelets scattered amongst them (Fig. 11.4c ).

As regards fabric, mortars can show fissures which cross the binder or move around the aggregate particles. These develop during mortar setting. Small pores can also be seen in the matrix (Fig. 11.5a ). Additives can modify the fabric, espe-cially if air-entraining agents are used. In this latter case, large, rounded pores are visible and impede the development of retraction fissures (Fig. 11.5b ).

11.4.2 Porosity and Pore Range Distribution

The fact that the porous system of a mortar has changed is not a conclusive result when it comes to assessing the progress of the carbonation process. The only thing it shows us is that the mineralogical changes somehow produce changes in the tex-tural properties of the mortar. In this case, a fall in the open porosity indicates that somehow the pores are closing. This is not surprising if we bear in mind that the volume of the calcite is higher than the volume of the portlandite. This increment corresponds to 11.8% (Lawrence 2005 ).

Fig. 11.4 SEM secondary images of mortar at different stages of carbonation: a at the beginning of the process; b during carbonation; c when carbonation is almost complete

20511 The Use of Lime Mortars in Restoration Work on Architectural Heritage

In general, a lime mortar with just a few days curing shows porosity levels of around 30–35%, while a mortar that has been curing for over a year shows falls in porosity to levels of around 25–28% (Cazalla 2002 ).

11.4.3 Elastic Parameters

The most interesting result shown by ultrasonic wave propagation is that after car-bonation the fabric has been altered. This is logical because a fall in porosity leads to an increase in the homogeneity of the sample. We also found that average veloci-ties in lime mortar specimens increased as time went by as a result of the carbona-tion process, from values of around 1800 m s 1 in the first few days of curing to values of around 2000 m s 1 after the first year (Cazalla 2002 ). The use of additives also affected the velocity of propagation of the ultrasound waves. The use of poz-zolana, for example, increases the ultrasound velocity of mortars (changing them into hydraulic mortars), whereas the air-entraining agent causes the velocity to fall, due to the development of air bubbles.

11.5 Conclusions

We can control and complete the process of mortar carbonation using different tech-niques. Our results show that carbonation is progressive and occurs from the outside in, gradually over time.

When it comes to selecting a high quality lime, we must take into account the raw material and the manufacturing process which must be carried out extremely carefully. This process must include the correct firing, slaking, storage and handling of the product. After the slaking process, the lime must not come into contact with the atmosphere as this would cause it to carbonate.

Fig. 11.5 Optical microscope photographs of a lime mortar without additives (a) and with air-entraining agent (b) . In the latter, round-shaped pores can be recognized

206 A. Luque et al.

Although previous studies indicate that mortars made with lime putty show higher levels of carbonation, our research shows that mortars made with lime pow-der can also reach comparable and even higher carbonation levels than those made with traditional lime putties. This may be due both to the quality of the raw material and to the carefully controlled manufacturing process through which mortars made with lime powder must pass.

The velocity of carbonation is a key parameter in any evaluation of the quality of a lime mortar. We have verified that once carbonate formation has reached high levels (i.e., 90%) under forced carbonation, the process slows down sharply. This may well be due to the great heat generated during the portlandite to calcite reac-tion, which causes capillary water to evaporate, and/or a high ambient temperature, which can reduce the solubility of the CO 2 , and/or the pore size reduction caused by calcite crystallization.

Under natural environmental conditions carbonation is a lot slower, and does not reach the same volume as that achieved by forced carbonation. We have dem-onstrated the importance of CO 2 concentration in the portlandite-to-calcite velocity reaction under certain levels of relative humidity and temperature. The reaction rate mostly depends on the reactivity of the lime and water content.

The addition of air-entraining agents to the mortars helps to eliminate or at least to reduce significantly the retraction fissures that develop in other types of mor-tars. This additive does not increase the carbonation rate of lime mortars because although round pores develop, there is a low interconnection between the pores.

Acknowledgements This research has been supported by the Research Projects MAT2004-6804 and 2004-CL019 and by the Research Group of the Junta de Andalucía RNM 179. We thank Nigel Walkington for the translation of the manuscript.

References

Arandigoyen M, Alvarez JI (2007) Pore structure and mechanical properties of cement-lime mor-tars. Cement and Concrete Research 37:767–775

Ashurst J (1990) Mortars for stone buildings. In: Ashurst J, Dimes GD (eds) Conservation of build-ing and decorative stone. Part 2, Elsevier, Amsterdam, pp 78–96

Bromblet P (1999) Properties and durability of air lime-based mortars for limestone repairs on monuments. RILEM international workshop on historic mortars: characteristics and tests, Uni-versity of Paisley, Scotland, p 11

C azalla O (2002) Morteros de cal. Aplicación en el Patrimonio Histórico. PhD thesis, Granada University, Spain

Cazalla O, Rodríguez Navarro C, Sebastián E, Cultrone G, De La Torre MJ (2000) Aging of lime putty: effects on traditional lime mortars carbonation. Journal of the American Ceramic Society 83:1070–1076

De Buergo Ballester MA, González Limón T (1994) Restauración de edificios monumentales. Monografías del Ministerio de Obras Públicas, Transportes y Medio Ambiente, Madrid, Spain

De La Torre MJ (1995) Estudio de los materiales de construcción en la Alhambra. PhD Thesis, Arte y Arqueología n. 28, Granada University, Spain

De La Torre MJ, Sebastián Pardo E, Rodríguez Gordillo J (1996) A study of the wall material in the Alhambra (Granada, Spain). Cement and Concrete Research, 26:825–839

20711 The Use of Lime Mortars in Restoration Work on Architectural Heritage

Dehilly RM, Tudo J, Sebai bi Y, Quéneudec M (2002) Influence of storage conditions on the car-bonation of powdered Ca(OH) 2 . Construction and Building Materials 16:155–161

Elert K, Rodríguez Navarro C, Sebastián E, Hansen E, Cazalla O (2002) Lime mortars for the conservation of historic buildings. Studies in Conservation 47:62–75

Furlan V, Bissegger P (1975) Les mortiers anciens. Historie et essais d’analyse scientifique. Revue suisse d’Art et d’Archéologie 32:1–14

Gaspar Tébar D (1996) Morteros de albañilería. Clasificación y propiedades. Degradación y con-servación del Patrimonio Arquitectónico. Curso de verano de El Escorial, Spain, pp 179–189

Hekal EE, Kishar E, Mostafa H (2002) Magnesium sulphate attack on hardened blended cement pastes under different circumstances. Cement and Concrete Research 32:1421–1427

Hendry EAW (2001) Masonry walls: materials and construction. Construction and Building Mate-rials 15:323–330

Hoffman P, Vetter G (1990) Analytical examination of mortars from medieval derelict village Holzheim. Fresenius Journal of Analitical Chemistry 338:133–137

Iglesias Martínez MC (1996) Análisis de la variación de la composición de los morteros utilizados en los muros de fábrica tradicionales: la compatibilidad de los morteros tradicionales de cal y la incompatibilidad de los morteros de cemento en el funcionamiento constructivo y estructural de los morteros de fábrica tradicionales. Actas del Primer Congreso Nacional de Historia de la Construcción. Madrid, pp 271–276

Lanas J, Álvarez Galindo JI (2003) Masonry repair lime-based mortars: factors affecting the mechanical behavior. Cement and Concrete Research 33:1867–1876

Lawrence RMH (2005) A critical review of techniques used to assess carbonation in lime mortars. International building lime Symposium. Orlando, Florida

Malinowski R (1981) Ancient mortars and concretes, durability aspects. Mortars, cements and grouts used in conservation for historic buildings. Proceedings of the Symposium, ICCROM, Rome, pp 341–350

Martín Pérez A (1990) Ensayos y experiencias de alteración en la conservación de obras de piedra de interés Histórico Artístico. Fundación Ramón Areces. Madrid, Spain

Martínez Ramírez S, Sánchez Cortes S, García Ramos JV, Domingo C, Fortes C, Blanco Varela MT (2003) Micro-Raman spectroscopy applied to depth profiles of carbonates formed in lime mortar. Cement and Concrete Research 33:2063–2068

Moorehead DR (1986) Cementation by the carbonation of hydrated lime. Cement and Concrete Research 16:700–708

Pérez Monserrat E, Baltuille Martín JM (2001) The rocks in monumental construction: deteriora-tion and techniques of intervention (Part 2: intervention techniques). Litos 55 (http://litoson-line.com/articles/55/ar5501s.shtml)

Radonjic M, Hallam KR, Allen GC, Hayward R (2001) Mechanism of carbonation in lime-based materials. Proceedings of the 8th euroseminar on microscopy applied to building materials, Athens (Greece), pp 465–75

Sbordoni Mora L (1981) Les materiaudx des enduits traditionnels. Mortars, cements and grouts used in conservation for historic building. Proceedings of the symposium, ICCROM, Rome, pp 375–383

Stamatakis MG, Papageorgiou A, Fragoulis AD, Chaniotakis E (2001) The nature of volcanic glass and its effect in the pozzolanic activity of tuffaceous rocks originated from Macedonia, north-ern Greece. Proceedings of the 8th euroseminar on microscopy applied to building materials, Athens (Greece), pp 271–280

Storemyr P (2004) Weathering of soapstone in a historical perspective. Materials Characterization 53:191–207

UNE-EN 459-1/AC (2002) Cales para construcción. Parte 1: Definiciones, especificaciones y cri-terios de conformidad. Madrid, Spain

Van Balen K (2005) Carbonation reaction of lime, kinetics at ambient temperature. Cement and Concrete Research 35:647–657

Van Balen K, Van Gemert D (1994) Modelling lime mortar carbonation. Materials and Structures 27:393–398